From a user centered point of view an important basic requirement to enable human-robot cooperation is to achieve conformity with operator's expectations of robot behavior. Therefore, this study focuses on the question, whether anthropomorphic robot movement trajectories can lead to an improved anticipation of the robot's behavior. Based on a virtual simulation environment a robotized assembly cell consisting of the assembly robot and the actual workplace was considered. In order to be able to simulate anthropomorphic movements, the human wrist trajectories of defined pick and place movements were obtained using an infrared motion capture system. The captured data were used to navigate the virtual assembly robot. Within the experiment anthropomorphic and robotic trajectories were distinguished. During the experiment, the main task of the participants was to predict the movement's destination as quickly as possible. Thus, the corresponding reaction value was analyzed to investigate the influence of anthropomorphic robot movements on human prediction in industrial environments.